A short (and slightly biased) history of collaborative editing
log.emonk.net
log.emonk.net
SubEthaEdit (while incredibly impressive and producer of a big technical feat at the time), is a _plain text_ collaborative editor.
Similarly, but higher up the scale, is Etherpad, which has the concepts of annotations for rich text.
On Google Docs and Google Wave, I can collaborate on rich text of arbitrary markup. For example, I can import any Word document and collaborate on it in real time. This is not a huge challenge from just the convergence / intention preservation algorithms perspective, but because they run on a browser. Wave went to the extent of normalizing the HTML representation of the `contenteditable` divs across all browsers to enable converging document results. Google Docs ditched `contenteditable` and rewrote an editor on top of plain HTML, with range selection logic and caret simulation built from scratch.
The fact that software like iWork Pages / MS Word[1] don't offer similar functionality (even with native, hand-made rendering engines and no browser support challenges) in 2011 speaks of Google's clear superiority in the office suite cloud offerings, which is the point the original article was trying to make.
[1] There's a plugin for Word http://www.codoxware.com/ that enables OT written by one of the most well-known OT researchers, who spoke at Google in '08 (http://www.youtube.com/watch?v=84zqbXUQIHc)
Topolsky seems to claim that collaborative editing with native apps is categorically not possible. That seems like a very odd claim given the history of collaborative editing.
Maybe he wanted to say that currently no native collaborative rich text editors exist (which is true), but he certainly failed to express that.
https://github.com/josephg/ShareJS
http://sharejs.org:8000/ (examples)
JSON OT is particularly neat for developing unique interactive applications, such as this board game: